Home | This Week's Contents  |  C&EN ClassifiedsSearch C&EN Online

Treatments have eluded drug developers to date, but the disease's multiple contributing factors suggest numerous therapeutic approaches

Alzheimer's Offers Small, 'Virtual' Company An Opportunity To Compete

Related Stories
Possible Amyloid Disease Therapy
[C&EN, Oct. 1, 2001]

Alzheimer Enzyme No Longer So Secretive
[C&EN, July 10, 2000]

Neuroscience: The Ultimate Moon Walk
[C&EN, Dec. 6, 1999]

Shedding Light On Amyloid Diseases
[C&EN, Apr. 5, 1999]

Related Sites
2000 Progress Report on Alzheimer's Disease

National Institute on Aging (NIA)

National Institutes of Health

University of Pennsylvania's Center for Neurodegenerative Disease Research


Bristol-Myers Squibb


Thomas Jefferson University

Elan Corp.

Andrx Corp.

Mindset BioPharmaceuticals

Immune Network

ReGen Therapeutics

Related People
Sangram S. Sisodia

Virginia M-Y. Lee

Jorge R. Barrio

Paul Greengard

D. Martin Watterson

Edward H. Koo

Mark H. Tuszynski

E-mail this article to a friend
Print this article
E-mail the editor
 Table of Contents
 C&EN Classifieds
 News of the Week
 Cover Story
 Editor's Page
 Government & Policy
  Government & Policy
 ACS News
 Digital Briefs
 ACS Comments
 Career & Employment
 Special Reports
 What's That Stuff?
 Pharmaceutical Century

 Hot Articles
 Safety  Letters

 Back Issues

 How to Subscribe
 Subscription Changes
 About C&EN
 Copyright Permission
 E-mail webmaster
March 11, 2002
Volume 80, Number 10
CENEAR 80 10 pp. 45-57
ISSN 0009-2347
[Previous Story] [Next Story]

Treatments have eluded drug developers to date, but the disease's multiple contributing factors suggest numerous therapeutic approaches


Enzyme inhibitors, chelation therapy, vaccines, anti-inflammatories, hormones, vitamins, exercise, and intellectually challenging activities: These are among the potential treatments under evaluation to ward off or even reverse the brain damage caused by Alzheimer's disease.

HELPING HAND Researchers are making progress on treatments that will help families cope with Alzheimer's disease.
Their diversity reflects the complexity and confusion surrounding the causes of this ravaging disease, which gradually erodes memory, learning and language skills, judgment, and the ability to perform even simple tasks. Patients eventually become completely dependent on others for care. Death generally follows diagnosis within a decade or less, though some sufferers may linger for 20 years.

Approximately 4 million Americans have the disorder, according to the National Institute on Aging (NIA), one of the National Institutes of Health. But the term Alzheimer's disease (AD) actually refers to more than one condition.

Early-onset, familial AD accounts for about 5% of cases and generally begins between the ages of 30 and 65. A child of a parent who has familial AD has a 25% chance of developing this form of the disease, which is caused by genetic mutations.

The brain lesions in these patients resemble those of somebody who is 75 or 80 years old and suffering from the more common form of the illness, known as sporadic AD. "And clinically, they have the same symptoms," says Sangram S. Sisodia, neurosciences professor and director of the Center for Molecular Neurobiology at the University of Chicago.

Sporadic AD usually develops after age 65. Its etiology is murky, but potential contributing factors include mutations or other genetic risk factors, head trauma, high cholesterol, heart disease, and high blood pressure.

"Geneticists are working hard to identify the genetic risk factors that would predispose an individual to AD," says Virginia M-Y. Lee, professor of pathology and laboratory medicine and codirector of the University of Pennsylvania's Center for Neurodegenerative Disease Research. "A good example is APOE (apolipoprotein E)," a gene that has three common forms (2, 3, and 4.) "If you have the APOE 4 allele, you are at higher risk to develop Alzheimer's disease." But the APOE 4 allele doesn't ensure that a person will get AD. "There are many of these risk factors to be identified, and we are just beginning to scratch the surface," Lee points out. "Basically, all we know with AD is that age is the most important risk factor, but we don't know why. And that may be because we don't know much about normal aging," she adds.

Though the familial and sporadic forms of AD differ, lessons learned from studying one form can be applied to the other, says Martin Citron, associate director for research at Amgen Neuroscience. The clinical picture and the pathology observed after death are very similar for both forms, he says, "so it seems justified to assume that the familial cases might be a model for what's going on in the sporadic cases." That means that although the causes of the two forms may be different, a treatment that intervenes in the pathogenic cascade of familial cases might also work for the sporadic cases.

Both types of AD are characterized by the deposition of disk-shaped clumps and fibrous bundles in patients' brains.

The clumps, known as amyloid or senile plaques, aggregate in the spaces between neurons. They contain insoluble snippets of amyloid- protein, as well as bits of neuronal and other cellular proteins.

"Basically, all we know with AD is that age is the most important risk factor, but we don't know why. And that may be because we don't know much about normal aging."

THE BUNDLES are composed of filaments of another protein, tau, that pair up in helices and form into neurofibrillary tangles within neurons. Normal tau stabilizes structures that shuttle vesicles containing neurotransmitters and synaptic proteins from cell bodies to nerve terminals. However, the tau in the tangles is an abnormal, excessively phosphorylated form of the protein that can't hold the neuronal shuttling structures together. The ensuing "collapse of the transport system first may result in malfunctions in communication between nerve cells and later may lead to neuronal death that contributes to the development of dementia," notes NIA in its "2000 Progress Report on Alzheimer's Disease" (http://www.alzheimers.org/pubs/prog00.htm).

Researchers are fairly certain that the tau tangles are neurotoxic. Sisodia points out, however, that these lesions are not restricted to AD, but also show up in patients with other neurodegenerative diseases such as Pick's disease and fronto-temporal dementias.

DEFENSE Eliminating amyloid formation early on may halt disease progression, Lee says.
AT THE SAME TIME, senile plaques may not be sufficient by themselves to cause AD, Lee notes. "Somehow, senile plaques cause the formation of neurofibrillary tangles," she says. "How, we don't know, but development of neurofibrillary tangles may be essential for the neuron to die in the brain of AD patients."

Most researchers believe that plaques are toxic and lead to neuronal dysfunction and death. But it's also possible that the plaques are less toxic than their constituent amyloid-b peptides, Sisodia says. "It could be the peptides themselves floating around in the brain and changing some aspect of neuronal cell biology that affects the synaptic interactions and synaptic transmission, leading to neurodegeneration."

In both familial and sporadic AD, amyloid- peptide is derived from amyloid- precursor protein (APP), a large molecule that pokes through the membrane of nerve cells. The enzyme -secretase--also known as beta-site APP-cleaving enzyme 1, or BACE1--initiates APP proteolysis. Much like a miniature shaver, BACE1 lops off part of the APP molecule on the outside of the membrane.

The enzyme -secretase--which may be a protein known as presenilin 1 (PS1) protein or may rely on presenilin to assemble into a functional complex--then cleaves what's left of the APP molecule in a region that traverses the membrane. That creates an amyloid- peptide (A)--consisting of 40 or 42 amino acids depending on where the cut is made--which is then released outside the nerve cell. A42 rapidly self-associates to form small "seeds" that serve as nuclei for accretion of more Ab peptides, Sisodia says.

A40 is more soluble and less liable to aggregate than A42. Healthy humans produce both forms in small amounts. In some familial AD sufferers, however, the ratio of A40 to A42 is strongly skewed in favor of A42 as a result of mutations in APP or presenilin. And although patients with sporadic Alzheimer's don't make excessive amounts of A, Citron observes, "they have tons of the material deposited in their brain plaques."

It's possible that amyloid builds up in AD patients because the mechanism that normally disposes of the peptide has broken down. Sisodia believes that over the next five years or so, studies will show that it's the ability to clear amyloid--either before or after it's deposited in brain plaques--that distinguishes the genes of some of those who get AD from those who don't. For instance, those who are genetically fortunate might produce beneficial forms of molecules such as -2-macroglobulin, urokinase plasminogen activator, and insulin-degrading enzyme that degrade Ab peptides, though this remains to be confirmed, he says.

TANGLED UP Neurofibrillary tangles and fiberlike processes of neurons (black) surround an amyloid plaque.
A third type of lesion, formed from aggregations of the protein synuclein, is also found in the brains of AD patients. These Lewy bodies can be detected in Parkinson's patients, too. Lee notes that the cause of their formation isn't yet understood, though it may be related to senile plaque formation.

The accumulation of multiple brain lesions in AD patients may lead to oxidative damage, brain inflammation, and eventually the loss of synapses, she adds.

Even as scientists wrestle with the hows and whys of AD etiology, they are moving ahead with potential treatments, though there is currently no cure on the market. A half dozen types of therapies are being explored, and most are aimed at related goals: blocking the formation or aggregation of amyloid or accelerating its clearance.

"If the genetics are telling us something, then by eliminating amyloid early enough--before you start the cascade of tangle formation, Lewy body formation, inflammation, and so on--you might have a chance of halting the disease," Lee explains. "If it's going to work, we'll know in the next five or 10 years. But if eliminating amyloid is not going to cure Alzheimer's disease, we'll have to go back to square one."

POTENTIAL THERAPIES are tested in mice genetically engineered to accumulate amyloid- plaques in their brains. Whether the results can be extrapolated to humans is open to debate, however, "because the way that we make these animals develop plaques is quite different than in humans," Lee says.

Sisodia adds that the mouse models are "spectacular in terms of amyloid deposition, and even for formation of neurofibrillary tangles. But the neurodegeneration that you see in Alzheimer's disease is not observed in these mouse models, with perhaps one exception."

And Citron points out that "cognitive testing in mice is difficult anyway, so the bottom line is there's no way to test the amyloid hypothesis other than having an antiamyloid drug in humans."

Researchers are doing their best to come up with some likely candidates.

Many major pharmaceutical companies are attempting to inhibit amyloid production with either -secretase or -secretase inhibitors. For instance, Bristol-Myers Squibb has a -secretase inhibitor in Phase I clinical trials.

Sisodia's group is trying to illuminate how presenilin, which is mutated in familial AD, functions and dysfunctions. "The question is, What does this molecule do normally?" he asks. "One thing that it does normally is play a very important role in development, in embryogenesis. It regulates the function of Notch, a receptor involved in cell fate decisions in early development."

In fact, its role is so crucial that knocking out the PS1 gene by genetically deleting it from mice causes the animals to die in utero, Lee says. In normal mice, presenilin sticks around even after development is complete, though its role in adulthood is unknown. So it's possible that researchers could avoid untoward side effects by inhibiting presenilin only in mature animals.

To test this possibility, Sisodia and his colleagues developed a mouse model in which expression of PS1 can be turned off at any time and in any site within the animal's body. They found that knocking out the presenilin gene specifically in the brains of adult mice enhanced memory storage, but its impact on AD is still uncertain [Neuron, 32, 911 (2001)].

Unlike the situation with presinilin, Lee says, knocking out the BACE1 gene creates mice that appear to be normal. "What's difficult is that there are no good inhibitors for BACE1 enzyme" that have been reported in the scientific literature, whereas "there's a whole host of inhibitors" for g-secretase, she says.

Citron, whose team identified the BACE1 enzyme [Science, 286, 735 (1999)], is trying to rectify that situation. That's tough to do, because inhibitors must be able to cross the blood-brain barrier. Also, "the active site is relatively large, and you have potential cross-reactivities with other proteases that you don't want to hit," he points out. That could lead to side effects.

Another therapeutic option is to allow production of Ab to proceed, but to disrupt its aggregation. One firm that has succeeded in doing that in mouse trials is Neurochem, which develops pharmaceuticals and diagnostic tools. The firm's proprietary Alzhemed compound mimics the glycosaminoglycan moieties of proteoglycans, which are found in amyloid plaques and stabilize and protect A fibrils from proteolytic degradation. Neurochem says Alzhemed interferes with that process by binding to A and blocking or delaying fibril formation. Preliminary results from a Phase I human clinical trial will be available in April.

Lee's lab is also investigating compounds to prevent plaque formation, though she has some concerns that such a strategy could keep potentially harmful A peptides in circulation if the brain isn't able to clear them out efficiently. She is working with small molecules based on histochemical dyes that bind amyloid.

Lee's team is also modifying some of these molecules so they can be used to image amyloid plaques in the brains of living patients [Proc. Natl. Acad. Sci. USA, 97, 7609 (2000)]. Such agents would allow researchers to determine if a particular therapy is effective by following changes in amyloid burden in the brain while a patient is being treated. Lee says most groups in the U.S. working in this area are using histological dye compounds that have been around for 50 or 100 years as a jumping-off point.

8010cov.brain 8010cov.brain
LOW METABOLISM PET scans show lower activity (blue and black regions) in the brain of an Alzheimer's patient (right) than in a healthy brain.

STARTING WITH Congo red and thioflavine S or T, Lee and her colleagues have developed some "promising radiolabeled compounds that bind to amyloid with high affinity," though so far they have only tested them in animal models. The compounds can cross the blood-brain barrier, which Lee notes the traditional dyes are unable to do.

Lee and her team were surprised that they could develop dyes that can differentiate between A40 and A42. That ability suggests that it may be possible to design molecules that can distinguish between amyloid plaques, neurofibrillary tangles, and Lewy bodies. "The first step is to identify the compounds, and then we need to radiolabel them efficiently and with high enough specific activity," Lee says. Depending on the radioactive isotope used, the dyes could be detected by positron emission tomography (PET) or single-photon emission computed tomography (SPECT).

University of California, Los Angeles, researchers, including molecular and medical pharmacology professor Jorge R. Barrio, recently succeeded at obtaining the first amyloid plaque images in living patients [Am. J. Geriatr. Psychiatry, 10, 24 (2002)]. They established that PET scans of the brains of patients injected with an 18F-labeled compound known as FDDNP, 2-(1-{6-[(2-[18F]fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malononitrile, can reveal plaques and tangles.

Such diagnostic tools would be welcome, since there is no test currently available--apart from autopsy--to conclusively diagnose AD. For now, doctors must rely on physical, psychological, and neurological exams and the patient's medical history.

Treatment with hormones including estrogen and testosterone is another potential AD therapy, but it's controversial. Many epidemiological studies have indicated that estrogen replacement therapy in postmenopausal women is beneficial, cutting risk of developing AD by about half, says Samuel E. Gandy III, director of the Farber Institute for Neurosciences and a professor of neurology, psychiatry, biochemistry, and molecular pharmacology at Thomas Jefferson University, Philadelphia. However, some other studies show no effect.

Lee speculates that "it may be beneficial to take estrogen before the onset of Alzheimer's. Once it develops, estrogen may not do anything. And that also seems to be the case for anti-inflammatories. They seem to be more efficacious when you take them prophylactically."

"If eliminating amyloid is not going to cure Alzheimer's disease, we'll have to go back to square one."

MISSING INGREDIENT Mouse models don't show the neurodegeneration typical of Alzheimer's disease, Sisodia says.
FURTHER EVIDENCE for a connection between sex hormones and AD comes from a study Gandy conducted of men undergoing hormone suppression therapy for prostate cancer [J. Am. Med. Assoc., 285, 2195 (2001)]. Once the patients' testosterone and estrogen levels had declined, their plasma amyloid levels roughly doubled. More data on estrogen's effects should begin to become available next year from rigorously controlled studies by investigators including Columbia University neurologist Mary Sano.

Gandy, who began his work in the field when he was a postdoc for Rockefeller University neuroscientist Paul Greengard, is trying to figure out how estrogen and testosterone modulate Alzheimer's risk. He believes the hormones inhibit amyloid accumulation. And they may do that by activating phosphorylation of a secretase regulator that breaks APP into harmless fragments. Gandy, Sisodia, Greengard, and their colleagues demonstrated that estrogen can diminish Ab generation in cultured neurons [Nat. Med., 4, 447 (1998)].

Gandy says pharmaceutical companies are trying to design selective estrogen receptor modulators that can duplicate the positive effects of estrogen while avoiding drawbacks such as cancer and venous thrombosis. Testosterone, too, is being considered as a starting point to develop analogs lacking its association with aggressive behavior, he notes.

Even if this route proves to be a dead end, there are several other options for treating amyloid. There's a Phase II clinical trial of the antibiotic clioquinol ongoing at the University of Melbourne in Australia by pathology professor Colin Masters and Ashley I. Bush, associate professor of psychiatry at Harvard Medical School. The drug is a chelator with a high binding capacity for zinc and some other metals--the same metals that help to aggregate amyloid peptides into plaques, Sisodia says. So the strategy is to use the drug to compete with amyloid for the metal ions, and as a result break apart the amyloid plaques. "And in fact, if you treat transgenic mice which have amyloid deposits with clioquinol, it clears the plaques," he says. The drug may also inhibit A-mediated production of damaging H2O2.

Immune therapy, initially developed by Elan Corp.'s Dale B. Schenk and colleagues [Nature, 400, 173 (1999)], is another antiplaque option that has attracted a lot of attention.

"The first study in 1999 showed a spectacular result," Lee says. Transgenic mice that had developed amyloid plaques were repeatedly vaccinated with synthetic A42. The treatment reduced or eliminated existing deposits and inhibited further plaque formation in the animals' brains. And Elan says that when it vaccinated young animals prophylactically, "virtually all of the mice had no detectable amyloid deposits" once they had aged.

The therapy is based on a "paradoxical finding, that immunizing with amyloid peptides results in clearance of amyloid plaques and even prevention of amyloid deposition in transgenic mouse models of Alzheimer's disease," Sisodia says. The therapy might work by prompting the immune system to recognize and attack the plaques within the brain, although some evidence indicates it might act as an A sink by moving A from the brain into the bloodstream.

The question is whether this would work in humans and whether clearing out plaque would result in an improvement in symptoms. In fact, the therapy had progressed to Phase II human clinical trials last year, but Elan suspended dosing this January after four patients out of about 360 developed serious central nervous system inflammation. Eleven other patients have since developed the same symptoms. The company is trying to determine what has gone awry.

Control of inflammation is the goal of some AD drug development efforts. The neuroinflammation characteristic of the disease is associated with excessive activation of glial cells and overproduction of cytokines and oxidative stress products. Northwestern University professor of molecular biology and biochemistry D. Martin Watterson and colleagues synthesized an alkylated 3-amino-6-phenylpyridazine derivative that inhibits these negative aspects of glial activation in vitro [J. Med. Chem., 45, 563 (2002)].

LONGTIME COLLABORATORS Gandy (right) and Norman R. Relkin, director of the Memory Disorders Program at Cornell University's Weill Medical College, review an Alzheimer's patient's APOE genotype and reduced brain activity as shown by a PET scan.
EPIDEMIOLOGICAL studies suggest that nonsteroidal anti-inflammatory drugs (NSAIDs) lower the risk of AD. And a trial in the Netherlands [N. Engl. J. Med., 345, 1515 (2001)] showed that NSAIDs such as diclofenac (about 100 mg per day) and ibuprofen (about 1,200 mg per day) that were taken before the onset of AD had a protective effect, says Lee, who takes ibuprofen and vitamin E prophylactically on a daily basis.

"How NSAIDs work, no one really knows," says Edward H. Koo, neurosciences professor at the University of California, San Diego. "The assumption was that the nonsteroidals reduce the inflammatory responses in the brain that accompany Alzheimer's." Although that may be true, it's not the whole story, he says. Koo, Todd E. Golde of the Mayo Clinic's neuroscience and pharmacology department, and their colleagues have determined that NSAIDs including ibuprofen, indomethacin, and sulindac sulfide actually reduce production of the toxic A42 peptide [Nature, 414, 212 (2001)]. The group also found that not all NSAIDs have this effect.

The work was carried out in tissue culture cells and in mice. Koo and the Mayo Clinic hope to start trials this year to find out whether these NSAIDs can cut A42 in humans as well. The clinic is also trying to design compounds that resemble these NSAIDs but lack the ability to inhibit cyclooxygenase (cox), the principal mode of action of these compounds. Cox catalyzes the formation of molecules that help initiate inflammation.

Another possibility is curcumin, a natural NSAID and antioxidant derived from the curry spice turmeric [J. Neurosci., 21, 8370 (2001)].

Numerous other treatments are being studied. Biotech company Immune Network, for instance, is carrying out a Phase II clinical trial to see if dapsone, an old leprosy drug with anti-inflammatory activity, has utility in slowing the progression of AD.

Nutritional supplements might also be helpful. Vitamin E, whose antioxidant properties may be important, could stave off the disease. Other antioxidants such as ginkgo biloba and Mindset BioPharmaceuticals' Oxigon (indole-3-propionic acid) are being studied. If the association between Alzheimer's and high blood levels of homocysteine [N. Engl. J. Med., 346, 476 (2002)] proves valid, it's possible that taking folic acid and vitamins B-6 and B-12, which lower homocysteine levels, could help.

And statins, which lower cholesterol, are being examined because cholesterol--or the mechanisms responsible for clearing it from the body--may be involved in the development of AD. Studies funded by pharmaceutical company Andrx Corp. indicate that lovastatin may lower amyloid- levels in the bloodstream, for instance.

Even frequent participation in cognitively stimulating activities might be beneficial [J. Am. Med. Assoc., 287, 742 (2002)].

UC San Diego associate neuroscience professor Mark H. Tuszynski and coworkers are conducting a Phase I clinical trial to test the efficacy of brain implants of a patient's own skin cells genetically modified to produce nerve growth factor. Animal studies have shown that such implants can prevent the death of the type of neuron that is extensively destroyed in human AD patients.

German pharmaceutical company Merz Pharma and its partners are near to marketing memantine to slow cognitive decline in AD. The compound is an N-methyl-D-aspartate receptor antagonist that may protect neurons from overstimulation caused by excess glutamate.

FOR NOW, clinicians must make do with compounds that moderate the symptoms of AD rather than alter the course of the disease. These include tacrine hydrochloride (Cognex), donepezil hydrochloride (Aricept), rivastigmine tartrate (Exelon), and galantamine hydrobromide (Reminyl). Such medications inhibit the metabolic breakdown of the neurotransmitter acetylcholine, which may improve an AD patient's thinking abilities and memory.

In all, these projects represent a massive investment in development of potential AD therapies. But a successful outcome can't be taken for granted. Koo, for one, doesn't see much chance of reversing AD once neurons have been destroyed. More promising, he believes, will be preventive measures taken before serious symptoms appear.

Even if some therapies prove out, it's going to take a while to find out. But that doesn't faze Lee. "In the next five or 10 years we should be able to come up with something to delay the onset or to retard the progression of AD," she predicts. "We have to be cautiously optimistic, because of the fact that we are--in all of these approaches--targeting amyloid. But the least we can accomplish in the next five to 10 years is to be able to test the amyloid cascade hypothesis, to see whether or not, if you eliminate amyloid, that you have delayed the onset or cured AD."

[Previous Story] [Next Story]


Chemical & Engineering News
Copyright © 2002 American Chemical Society

Home | Table of Contents | News of the Week | Cover Story
Business | Government & Policy | Science/Technology
Chemical & Engineering News
Copyright © 2002 American Chemical Society - All Right Reserved
1155 16th Street NW • Washington DC 20036 • (202) 872-4600 • (800) 227-5558

CASChemPortChemCenterPubs Page